The scaling of a T- or F-contrast vector has no effect on the resulting
T or F statistic (because both numerator and denominator of the
statistic are scaled the same way).
However, be careful when you plot the contrast of parameter estimates
associated with that contrast in SPM: this contrast (eg numerator of a
T-statistic in the case of a T-contrast) will depend on the scaling,
corresponding either to the sum or the average across conditions in your
two example contrast vectors below.
Rik
Chris Watson wrote:
> Hi Chris et al,
> I just tested this out on one of our datasets. Usually we have a
> contrast [1 0 1 0 1 0] to test for an average effect; I tried [1/3 0
> 1/3 0 1/3 0], and got the same exact result. Perhaps SPM accounts for
> contrasts such as these, somehow? I think the answer is around lines
> 95-100 in spm_contrast.m...
>
> Chris
>
> Benjamin, Christopher wrote:
>> Hi, related question-if you want to test the average slope/effect
>> don't you need to weight each variable by 1/(# of variables)? I
>> thought if you weight each column of interest by 1 you're effectively
>> summing the beta for each? Then And test the average effect against
>> the baseline, if your baseline is explicit you'd weight that by -1,
>> but if your baseline's implicit you wouldn't have any negative values?
>>
>> Thanks for the clarification!
>>
>> On 29/01/2010, at 10:12, "MCLAREN, Donald"
>> <[log in to unmask]<mailto:[log in to unmask]>> wrote:
>>
>> Assuming that these are both continuous variables, it means that the
>> slope of the relationship between X1 and Y is greater than the
>> relationship between X2 and Y. X1 has 1 over it and X2 has a -1 over
>> it. You can only tell +/- correlation if you have a 1 or -1 over a
>> single column OR if you want the average slope you can have 1s or -1s
>> over several columns, but you can have both. As soon as you have
>> both, then you are testing the difference in the slopes.
>>
>> Best Regards, Donald McLaren
>> =================
>> D.G. McLaren
>> University of Wisconsin - Madison
>> Neuroscience Training Program
>> Office: (608) 520-0586
>> =====================
>> This e-mail contains CONFIDENTIAL INFORMATION which may contain
>> PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED
>> and which is intended only for the use of the individual or entity
>> named above. If the reader of the e-mail is not the intended
>> recipient or the employee or agent responsible for delivering it to
>> the intended recipient, you are hereby notified that you are in
>> possession of confidential and privileged information. Any
>> unauthorized use, disclosure, copying or the taking of any action in
>> reliance on the contents of this information is strictly prohibited
>> and may be unlawful. If you have received this e-mail
>> unintentionally, please immediately notify the sender via telephone
>> at (608) 520-0586 or email.
>>
>>
>> On Fri, Jan 29, 2010 at 3:03 AM, Kelly Atalaia Silva
>> <<mailto:[log in to unmask]>[log in to unmask]<mailto:[log in to unmask]>>
>> wrote:
>> Hello,
>> I would like to know how can I know the direccion of one
>> correlation in SPM. For example, if I put "1" on one variable of
>> interest and "-1" into another variable does it means that there is a
>> negative correlation between both variables?
>> Thanks in advance,
>> Kelly
>>
>>
--
-------------------------------------------------------
Dr Richard Henson
MRC Cognition & Brain Sciences Unit
15 Chaucer Road
Cambridge
CB2 7EF, UK
Office: +44 (0)1223 355 294 x522
Mob: +44 (0)794 1377 345
Fax: +44 (0)1223 359 062
http://www.mrc-cbu.cam.ac.uk/people/rik.henson/personal
-------------------------------------------------------
|